Melanoma is the most deadly type of skin cancer with a five-year survival rate for late stage cancer of only 15%. Key reasons for the poor prognosis of this tumor type are its high rate of metastasis and resistance to therapy. To date, very little is known about the molecular pathways governing the acquisition of the metastatic phenotype and therapy resistance in melanoma. Emerging evidence suggests that cancer cells use developmental programs for tumor dissemination, propagation, and resistance to currently available therapies. Our overall hypothesis is that melanoma cells hijack the neural crest developmental program for their progression. We recently discovered that neural crest-like reprogramming is achieved in melanocytes when Msh homeobox 1 (MSX1) is overexpressed. MSX1-expressing melanocytes exhibit a similar morphology and phenotype to neural crest-like precursor cells in human skin. In the embryo, MSX1 is a transcriptional repressor essential for neural crest development. Our data demonstrate that MSX1 binds and suppresses the proximal promoter of microphthalmia-associated transcription factor (MITF), the master transcriptional regulator of melanocyte differentiation. Expression of MSX1 correlates with melanoma progression, and our data show that knockdown of MSX1 in metastatic cells significantly impairs motility in vitro and reduces liver colonization in vivo. MSX1 expression is upregulated by inflammatory factors in the microenvironment, which has been shown to play a role in cellular plasticity and immunotherapy resistance in melanoma. In this proposal, using our knowledge of tumor biology and tumor immunology, we propose to study a novel link between the embryonic phenotype of cancer and immune response to cancer cells. Our goal is to delineate the mechanisms for how MSX1 dictates a neural crest-like plastic phenotype in melanoma cells, which helps them escape immune surveillance, and ultimately promotes their metastasis. In Specific Aim 1, we will define the role of neural crest-like reprogramming in resistance to immunotherapy. We will take gain-of-function (cDNA over-expression) and loss-of-function (RNAi) approaches using lentiviral vectors to assess the biological significance of MSX1-induced neural crest-like dedifferentiation in human melanoma cells using the cytotoxic T lymphocyte assay and an immune competent, syngeneic, allograft mouse melanoma system. We expect that by targeting neural crest-like reprogramming, this will freeze the phenotype of melanoma cells in more differentiated and more immunogenic state, thereby sensitizing melanoma cells to currently available immunotherapies. The design of small molecules inhibiting transcription factors is a challenge; therefore we will explore upstream inducers and downstream targets of MSX1 to identify melanoma-specific druggable targets for MSX1-driven dedifferentiation in Specific Aim 2. Successful completion of this study will identify melanoma-specific dedifferentiation mechanisms and provide a proof-of-principle to target cellular plasticit in melanoma, especially in combination with immunotherapy.